Surgery versus SABR for resectable non-small-cell lung cancer - Part 4 of 9

This is a letter in response to the article by Chang and colleagues. The authors point out that only 58 patients were enrolled from 38 centers over 66.3 months, and speculate that this might be due to patients' preference for surgery. Since the cohort was small, follow-up period brief, and analyses retrospective in nature without power calculations, at best these results may generate hypotheses. First, the conduct of the two randomised controlled trials (RCTs) was different and this was compounded by variation among the treatment centers within each trial. Second, a large proportion of patients in the surgical arm (3 of 27, 11%) did not undergo the intended treatment, lobectomy. Third, ROSEL trial patients did not require histologic confirmation of NSCLC prior to randomisation. Rather, patients with FDG avid lesions on preoperative PET scans were included; however, there were key differences in how PET scans were obtained at various centers. Fourth, only 5 of 27 patients in the surgical arm underwent VATS lobectomy, the majority underwent thoracotomy. The authors state that certain complications are less likely after VATS lobectomy compared with thoracotomy and that this approach is associated with similar and possibly better long-term survival than an open approach. They conclude by disagreeing that SABR is associated with better overal survival, concurring that SABR may be better tolerated than surgery, and express that further study of VATS lobectomy and subanatomic resection with lymph node assessement versus SABR and open thoracotomy are needed.